IUBio Biosequences .. Software .. Molbio soft .. Network News .. FTP

No subject


Sun Apr 10 21:14:20 EST 2005


several transformations). The values of these statistical parameters are with statistical methods
(T-probe) compared to the values of corresponding paramters of a collection of 'normals', the
reference file.
The user of the neurometrics method interprets the differences and uses it as an extra argument for
his diagnosis.

The science behind the Neurometrics method has two 'schools':

- Dr. Roy John
- Duffy.

The preparation and collection of the parameters has the following steps:
Step 0: select a person to be examined with Neurometrics
Step 1: record at least 100 seconds of clean EEG
Step 2: make lists of parts of the EEG suitable to process  with the method
Step 3: execute spectral analyses on the selected EEG parts.

Neurometrics uses multivariate analyses. The parameters are extracted with a computer from the EEG
spectra in the following steps (see John83 page 254):

1. Quantitative univariate features are extracted from the EEG data by computer calculations.
2. Adequate transformations to get Gaussian distributions are applied to every type of univariate
feature.
3. Age-regression curves are calculated for the mean and standard deviation of every Gaussian-
transformed univariate feature.
4. Z-transformations are calculated for the univariate features from every sample, the resulting Z-
values per feature appear to be distributed Gaussian.
5. Multivariate features are calculated. The Mahalanobis distance is used to correct for the cross-
correlation of the features.
6. Adequate transformations to get Gaussian distributions are applied to every Mahalanobis distance.
7. Age-regression curves are calculated for the mean and standard deviation of every Gaussian-
transformed Mahalanobis distance (like step 3 but now for the multivariate features).
8. Z-transformations are calculated for every Mahalanobis distance from every sample (like step 4
but now for the multivariate features). The resulting Z-values per Mahalanobis distance appear to be
distributed Gaussian.

The features taken in account are
Type of variable
· original signal,
· asymmetry,
· coherence
Locations	
· Central
· Temporal
· Parieto-Occipital
· Fronto-Temporal
Skull part	
· left,
· right	
Frequency band	
· alfa Powerspectrum,
· beta Powerspectrum,
· delta Powerspectrum,
· theta Powerspectrum,
· low frequency Powerspectrum
· total Powerspectrum


  SOME INFORMATION ON CHAOS THEORY-BASED EEG-RESEARCH
  ---------------------------------------------------
  Being not a neuroscientist I think best I can do is cite the abstracts from  accepted
  scientific articles on the subject.

  [STAM94]:  "  The irregular, aperiodic character of the EEG is usually explained by a
  stochastic model. In this view the EEG is linearly filtered noise. According to chaos
  theory such irregular signals can also result from low dimensional deterministic chaos. In
  this case the underlying dynamics is nonlinear, and has only few effective degrees of
  freedom. In contrast, stochastic models are less efficient, because they require in
  principle infinite degrees of freedom. Chaotic dynamics in the EEG can be studied by
  calculating the correlation dimension (D2). Although it has become clear that D2
  calculations alone cannot prove chaos, the D2 has potential value as an EEG diagnostic. In
  this study we investigated whether D2 could be used to discriminate EEGs from normal
  controls, demented patients and Parkinson patients. We have analyzed epochs (20 channels;
  2,5 s) from 52 EEGs (20 controls; 15 patients with dementia; 17 patients with Parkinson's
  disease). Controls had a mean D2 of 6.5 (0.9); demented patients of 4.4 (1.5), and
  Parkinson patients of 5.3 (0.9). Both groups were significantly different from controls
  (p<0.001). There was a significant positive correlation between D2 and relative power in
  the beta band (r=0,81) and a significant negative correlation between D2 and power in the
  delta (r=-0.60) and the theta band (r=-0.37). These results suggest the possible
  usefulness of multichannel D2 estimations in a clinical setting.   "

  [JELL95]  "  .... As the definite diagnosis of Alzheimer disease depends upon pathologic
  examination, in vivo diagnostic tools are of great importance. This becomes even more
  urgent now that new therapies such as tacrine are available. Neuroimaging procedures are
  particularly valuable for the exclusion of certain causes of dementia and Parkinsonism.
  Neurophysiologic tests, especially the EEG, have been examined for their diagnostic value
  in Alzheimer disease and Parkinson disease in AD, the EEG shows well documented
  abnormalities in the course of the disease. With spectral analysis, decrease of alpha and
  beta activity and increase of delta and theta activity are found.
  .....
  Another approach to quantification of EEG abnormality is to determine the degree of
  desynchronization of the EEG. Desynchronization is associated with cholinergic activation
  of the cortex, in particular during mental activity. Desynchronization can be quantified
  with several techniques. Recently we described a new measure (Acceleration Spectrum
  Entropy, ASE) to quantify EEG desynchronization. Usefulness of the ASE was investigated in
  normal EEGs during different activational states. The ASE was found to be a simple and
  sensitive measure of desynchronization. In a second study we showed that the ASE is
  sensitive to localized changes in desynchronization related to mental activity in healthy
  persons.
  In this study, we measured the ASE in patient groups with dementia (AD) and PD. The aim of
  our study was to determine whether ASE is a useful instrument for differentiating patient
  groups from each other and from normals. Furthermore, we hoped to find certain patterns of
  desynchronization in the AD and PD patient groups that could help in choosing therapeutic
  strategies.   "


  REFERENCE LIST
  --------------
  JOHN83   E.R. John, L. Pricher, H. Ahn, P. Easton, J. Fridman and H. Kaye;
         " Neurometric evaluation of cognitive dysfunctions and neurological
           disorders in children "
       In: Progress in neurobiology Vol 21, pp 239 to 290; 1983 Pergamon Press Ltd.

  JOHN90   E. Roy John, Ph.D.;
         " Principles of Neurometric"s  "
       In: Am. J. EEG Technol. 30 251-266, 1990 IOWA U.S.A.

  JONK93   E.J. Jonkman;
         " Wie is normaal? "
       In: Symposium Klinische Neurofysiologie: "Computeranalyse van het EEG";
           3-11-1993 VU Amsterdam,

  WEER93   A.W. de Weerd;
         " Neurometrics: groeps- en individuele diagnostiek  "
       In: Symposium Klinische Neurofysiologie: "Computeranalyse van het EEG";
           3-11-1993 VU Amsterdam,

  STAM94   Stam, Kees J.; Tavy, D‚nes L.J.; Jelles, Brechtje; Achtereekte,
           Herbert A.M.; Slaets, Joris P.J.; Keunen, Ruud W.M.
         " Non-Linear Dynamical Analysis of Multichannel EEG: Clinical
           Applications in Dementia and Parkinson's Disease "
       In: Brain Topography, Volume 7, Number 2, page 141-150, (1994)
       Corresponding author: Dr. Kees J. Stam, Department of Geriatrics,
           Leyenburg Hospital, P.O.Box 40551, 2504 LN  The Hague, The Netherlands.

  STAM95   Stam, C. J.; Jelles, B.; Achtereekte, H.A.M.; Rombouts, S.A.R.B.;
           Slaets, J.P.J.; Keunen, Ruud W.M.
         " Investigation of EEG non-linearity in dementia and Parkinson's
           disease "
       In: Electroencephalography & Clinical Neurophysiology 95 page 309-317 (1995)
           Official Organ of the International Federation of Clinical
           Neurophysiology, published by Elsevier Science Ireland Ltd
       Corresponding author: C.J. Stam Tel +31 70 3592000

  JELL95   Jelles, B.; Achtereekte, H.A.M.; Slaets, J.P.J.; Stam, C. J.
         " Specific Patterns of Cortical Dysfunction in Dementia and
           Parkinson's Disease Demonstrated by the acceleration Spectrum
           Entropy of the EEG "
       In: Clinical Electroencephalography, Volume 26 Number 4, page 188-192 (1995)
       Corresponding author: Dr. C.J. Stam, Department of Neurology and Clinical
           Neurophysiology, Leyenburg Hospital, Leyweg 275, P.O.Box 40551,
           2504 LN  The Hague, The Netherlands.


  ABSTRACTS OF PRECEDING POSTINGS ON PERSONHOOD AND EEG
  -----------------------------------------------------
  John Parrish THOMPSON <gsi03919 at gsaix2.cc.GaSoU.EDU> december 1995 wrote:
  > I am currently working on a theory of personhood and its defining
  > characteristics and desperately need expert information on comparative
  > brain wave patterns between "normal" conscious humans, sleeping humans,
  > higher order primates, such as chimpanzees, mentally handicapped
  > children, comatose humans (not braindead), and human fetuses.
  .....
  > In a move of extreme reductionism I am seeking to isolate one
  > defining characteristic which would conclusively determine the personhood
  > or nonpersonhood of the subject in question. I believe to this
  > may lie in human brainwave patterns.

  David SEAMAN <ds005c at UHURA.CC.ROCHESTER.EDU> on 28 december 1995 answered:
  > You're right, this is extreme reductionism.
  ...
  > You're disregarding the fact that "person" is a concept, word, or what
  > have you, that has been created by humans. It has no relevance to the
  > real world other than how it is defined.
  > Not to mention that it is fairly ridiculous to try to find any
  > abstract characteristic easily identifiable in an EEG. At least, if there
  > is such a pattern, you really couldn't find it by comparing monkeys,
  > fetuses, and adult humans. There are simply going to be differences in
  > these organisms'EEG's as a result of their biology, or perhaps as a result
  > of their biology, there will be no differences. But that couln't really
  > come close to defining "personhood" for you - you'd have to say, "Oh, there
  > are some differences between a monkey's EEG and a human's. Well, I'm going
  > to define personhood as having the characteristic human EEG." The problem
  > is, again, that "personhood" is an arbitrary categorization of certain
  > levels of complexity in brain organization, function, and structure. There
  > may be correlations, but all it would confirm is that we, the arbitrators,
  > had some intuition about certain things being complex in brain function,
  > and some things not being as complex as we are.

  John Parrish THOMPSON <gsi03919 at gsaix2.cc.GaSoU.EDU> furthermore contributed:
  > Is it your assertion that an analysis of EEG patterns by either a
  > neuroscientist or, perhaps, a computer analysis could not differentiate
  > a single pattern, frequency, or wave, common only to humans
  > and not present in higher order primates? They would
  > appear identical even to a skilled professional?

  Mariela SZIRKO <postmaster at neubio.sld.ar> on 4 Jan 1996 continued:
  > .. the conceptual error I pointed is in trying to apprehend it [personhood]
  > in general,
  ...
  > Certainly, but it [the attempt to apprehend the traits of personhood as used
  > by our society to determine the rights of entities} is a sociological study;
  ...
  > Sure enough. EEG and MEG give notice, blurred within certain gross
  > time resolution, of the speed of changes in the "instantaneous" state
  > of the distribution of charges accross the organic tissue volume. This
  > is a part of brain psychogenetic Function 1. Functions 2 and 3 are
  > not probed, albeit they are central for giving contentsto consciousness.
  > Moreover, besides contents, consciousness (experiencing) is a general
  > physical function.
  ...
  > So there is no reason to expect that one or several patterns of Function 1
  > could distinguish the psychogenetic processes of one species from those of
  > others, specially inside the same order. To try distinguish them we should
  > first have described the psychogenetic physical processes,
  ...
  > then we should begin to apply the so acquired concepts to compare very
  > different psychogenetic processes ..

  Eveline BERNARD <ebernard at pi.net> (that's me) wrote:
  > I agree with the Argentinian lady that it can be dangerous to try to connect
  > personhood to EEG brainwave patterns.
  > Nevertheless I think it is possible to find significant differences in EEG
  > brainwave patterns between groups of individuals.
  > Look for litterature on NEUROMETRICS (dr. Roy John, U.S.A., and dr. de Weerd,
  > Westeinde Ziekenhuis, den Haag, The Netherlands) and CHAOS THEORY (Research
  > in the Leyenburg Ziekenhuis, Den Haag, The Netherlands).
  > The Neurometrics method is based on multivariate analyses, a statistical technique.
  > It finds differences between e.g. man and woman, younger and older people.
  > The Chaos method finds differences between e.g. more or less demented people.

  Mariela Szirko <postmaster at neubio.sld.ar> on 4 Jan 1996 reacted:
  > Can we ascertain on a single EEG if it pertains to a man or a woman, a younger
  > or older human (and then and provided some further comparations, if it pertains
  > to another species?
  ...
  > While I can usually distinguish among vigilance states, I confess to have never
  > performed such sophisticated reshuffling of EEG values. It might be an error due
  > to my following prejudice: the EEG data already convey cancellations and additions
  > of so multiple sources that I presuppose that further noise (added in reduction)
  > will induce false patternings.  So we don't trespass the conventional tools of
  > mathematical analysis usual in brain mapping, frequency weight and spectral
  > studies.  I always bear in mind the physicochemical sources of the EEG, rather
  > than emphasizing its clinical practicalities.
  ... [Szirko gives anecdote about bad EEG research in the far past]
  > However probably the newly-developed methods Eveline mentions can pick up real
  > facts on INDIVIDUAL EEG.  Which ones? How much? How reproducibly?
  > So let me inquiry: ASIDE FROM GROUP DIFFERENCES, is it already feasible with
  > those new methods to establish with reasonable confidence the sex, age, species
  > or other non-EEG feature of one concrete EEG?

Disclaimer: The opinions in this message are personal. The writer does not accept any
responsibilty for damage caused by reading or interpreting the contents.

=====================   ===============================
Name:                   Company:
 Eveline Bernard         Ir. E.D.A.S. Bernard RI
Address 1:              Address 2:
 Stationsweg 56          Paalhoeveweg 22
 2515 BP  DEN HAAG       3775 KL  KOOTWIJK
 The Netherlands         The Netherlands
Tel +31 (0)70 3884561   Tel +31 (0)577 456543
Mail address:           Mail alias:
 ebernard at pi.net         ir._e.d.a.s._bernard_ri at pi.net
=====================   ===============================





More information about the Neur-sci mailing list

Send comments to us at biosci-help [At] net.bio.net